Kortix

Read and write real office files: PDF, Word, Excel, and presentations.

Everything for evidence-based research: deep research, cited reports, and academic paper search.

Outbound-ready: account research, outreach drafting, call prep, and competitive analysis.

Answer the user's questions about what Kortix itself can do — its capabilities and how they fit together: research and the web, browser automation, code and data execution, document and media creation, building and deploying websites and apps, connectors and integrations, secrets, memory, scheduling and triggers, channels, parallel subagents, and model selection. Use whenever the user asks 'what can you do', 'can you do X', 'how does Kortix work', 'how do I do Y in Kortix', or compares Kortix to other AI tools.

Get a complete picture of any company or person before outreach. This skill always works with web search, and gets significantly better with enrichment and CRM data.

Browser automation skill for using the baked agent-browser runtime from an OpenCode agent.

Score and rank candidate résumés against one job and hand back an interview shortlist. Use when you already have applicants plus a role and want them triaged by screen-in likelihood — 'rank these CVs for the Staff Backend role', 'who should we interview from this requisition', 'shortlist the applicants in our ATS for this opening'. Pulls applications and résumés from a connected applicant-tracking system, or scores résumé files you upload. Not for finding/sourcing new candidates and not an auto-reject gate.

SOX 404 control testing methodology, sample selection approaches, testing documentation standards, control deficiency classification, and common control types.

Run a recurring outbound sourcing workflow for a role: source candidates, dedupe against your ATS pipeline, rank them, and serve a review surface where a recruiter marks Yes/Maybe/No — then feed those decisions into the next run. Use to operate, schedule, debug, or rebuild a standing sourcing pipeline ('do this week's sourcing run for the Platform role', 'set up recurring sourcing', 'regenerate the candidate review app'). Not for one-off people lookups — use people-search for those.

Frameworks for documenting, applying, and enforcing brand voice and style guidelines across marketing content.

Get fully prepared for any sales call in minutes. This skill works with whatever context you provide, and gets significantly better when you connect your sales tools.

Frameworks and guidance for planning, structuring, and executing marketing campaigns.

You are a response template assistant for an in-house legal team. You help manage, customize, and generate templated responses for common legal inquiries, and you identify when...

Month-end close checklist, task sequencing and dependencies, status tracking, and common close activities organized by day.

Route code and repository work to isolated Kortix sessions instead of burning your main context on it. Use when asked to implement a feature, fix a bug, make failing tests pass, work a ticket/issue, refactor, navigate or change a codebase, ship a PR, or review a pull request — and when the user wants two agents or parallel reviewers on the same code. Covers the explore-vs-session routing call, deciding where the code lives (this project's repo vs cloning an external GitHub repo), finding the repo with gh, spawning sessions in parallel, dual PR reviews, and reporting results back. Triggers: 'implement', 'fix the bug', 'make the tests pass', 'work this ticket', 'refactor this', 'open a PR', 'review this PR', 'run two agents on it'.

Use for routing work that mixes coding, repository implementation, SQL/data analysis, or technical investigation and should often be delegated to a background session with clear context and boundaries.

You are an expert at competitive analysis for product managers. You help analyze competitors, map competitive landscapes, compare features, assess positioning, and derive strate...

Research your competitors extensively and generate an interactive HTML battlecard you can use in deals. The output is a self-contained artifact with clickable competitor tabs an...

You are a compliance assistant for an in-house legal team. You help with privacy regulation compliance, DPA reviews, data subject request handling, and regulatory monitoring.

Guidelines and frameworks for creating effective marketing content across channels.

You are a contract review assistant for an in-house legal team. You analyze contracts against the organization's negotiation playbook, identify deviations, classify their severi...

Use for building prospect-specific sales assets such as landing pages, decks, one-pagers, and workflow demos.

Author or edit an Agent Skill that follows the agentskills.io spec — the SKILL.md frontmatter and body, plus optional scripts/, references/, and assets/. Use when the user wants to create a new skill, turn a repeatable workflow into reusable know-how, edit an existing skill (its name, description, instructions, or any frontmatter field), restructure or split a skill, validate one, or package one for sharing or for the Kortix marketplace.

You are an expert at conducting multi-source research to answer customer questions, investigate account contexts, and build comprehensive understanding of customer situations. Y...

Get a clear view of what matters most today. This skill works with whatever you tell me, and gets richer when you connect your tools.

Deep research agent skill. Use when the user needs thorough, scientific, truth-seeking research on any topic -- investigating claims, finding primary sources, synthesizing evidence, producing cited reports. Triggers on: 'research this', 'investigate', 'deep dive', 'find sources', 'what does the evidence say', 'literature review', 'fact check', 'analyze the research on', any request requiring multi-source investigation with citations.

Artifact-agnostic design guidance — works for CSS, PowerPoint, matplotlib, PDF, or any visual output.

Use for structured review, fact-checking, annotation, and correction of documents such as PDF, DOCX, PPTX, and XLSX files.

Use for creating, editing, extracting, converting, and reviewing Word documents.

Free domain research and availability checking. No API keys or credentials required. Uses RDAP (1195+ TLDs) with whois CLI fallback for universal coverage. Checks if domains are available, searches keywords across TLDs, performs WHOIS/RDAP lookups, checks expiry dates, and finds nameservers. Use when the agent needs to: check if a domain is available, search for domains, find who owns a domain, check domain expiration, get nameservers, bulk check domains, or do any domain research. Triggers on: 'check domain', 'is domain available', 'search domains', 'domain availability', 'who owns this domain', 'whois', 'domain expiry', 'when does domain expire', 'nameservers for', 'domain research', 'find domains for', 'domain ideas', 'bulk domain check'.

Research first, then draft. This skill never sends generic outreach - it always researches the prospect first to personalize the message. Works standalone with web search, super...

Read and analyze a connected electronic health record — lab results and blood work, current and past medications, diagnosed conditions, allergies, procedures and surgeries, immunizations, appointments, and patient summaries. Use when the user asks 'what were my latest labs', 'what medications am I on', 'what's my medical history', 'am I up to date on vaccines', 'what am I allergic to', or wants a clinical summary. Marketplace skill — requires a connected EHR / medical-records provider (e.g. a FHIR-based health system or lab platform).

ElevenLabs audio generation — text-to-speech, voice cloning, and sound effects. Use this skill any time the agent needs to: convert text to spoken audio, narrate documents or content, generate voiceovers, clone voices from audio samples, create sound effects, or produce any audio output from text. Supports multiple voices, languages, models, voice cloning, batch processing, and sound effect generation. Requires ELEVENLABS_API_KEY.

High-precision people and company lookup backed by a Kortix people-data connector (enrichment, prospecting, or contact-data provider such as Apollo, People Data Labs, Clearbit, Clay, RocketReach). Use when such a connector is configured and you need verified profiles, work emails, phone numbers, employment history, or org charts — or to enrich an existing list of names, domains, or profile URLs into full records. When no people-data connector is present, hand off to the web-search people-search skill instead.

You are an expert at determining when and how to escalate support issues. You structure escalation briefs that give receiving teams everything they need to act quickly, and you...

Systematic methodology for profiling datasets, assessing data quality, discovering patterns, and understanding schemas.

You are an expert at writing product requirements documents (PRDs) and feature specifications. You help product managers define what to build, why, and how to measure success.

Public-market data and analysis: live and historical stock/ETF/crypto/index prices, market cap, valuation multiples, company financials (income statement, balance sheet, cash flow), earnings transcripts and estimates, analyst targets, dividends, debt, M&A, sector peers, ETF holdings, insider and institutional activity, private/pre-IPO valuations, and macro indicators (GDP, inflation, rates). Also covers filing-grounded facts — a company's officers, board, executive compensation, deal terms. Use for any question whose answer is a current or point-in-time market number, never a number from memory.

Formats, GAAP presentation requirements, common adjustments, and flux analysis methodology for income statements, balance sheets, and cash flow statements.

Use when the user wants a YouTube transcript from a single URL or video ID. Optimized for one input and one output: fetch the transcript fast, default to plain transcript text only, and avoid extra commentary unless the user asks for timestamps, JSON, or metadata. Triggers on: youtube transcript, transcript from this video, get captions, extract transcript from YouTube, summarize this YouTube transcript after fetching it.

Directs image-generation requests — decides the capability the image needs, writes the prompt, assigns reference-image roles, and holds the safety line over the replicate skill. Engage when a request needs a routing decision (transparency, legible text/UI, in-place edit, premium polish, fast ideation), prompt shaping, or a safety judgment (real people, brands, minors, evidence-like imagery). This skill directs; it does not generate.

Migrate a user's local Claude Code / Codex setup into Kortix — Agent Skills, memories/rules, and MCP server connectors. Use when the user wants to bring their local Claude Code or Codex skills, CLAUDE.md / AGENTS.md preferences, or configured MCP servers over to a Kortix project, or says 'import my local skills', 'bring my Claude memories over', 'match my MCPs to Kortix connectors', 'set up what I use locally here', or 'switch from Claude Code / Codex'.

Structured, multi-step investment research: screen and rank stocks against criteria, stress-test a bull/bear investment thesis, evaluate a name through a famous investor's philosophy (Buffett, Lynch, Druckenmiller, Soros, Dalio, Ackman, Burry, Wood), analyze a portfolio's diversification and risk, or backtest a historical strategy. Use when the request is a workflow — screening, scoring, risk analysis, backtest, or a formal thesis pressure-test — not a one-line buy/sell read.

Best practices, standard entry types, documentation requirements, and review workflows for journal entry preparation.

You are an expert at creating, organizing, and maintaining support knowledge base content. You write articles that are searchable, scannable, and solve customer problems on the...

>-

How to reach a CONNECTED MACHINE (a user's laptop/desktop, or any computer paired over the Agent Computer Tunnel) from a Kortix session — read/write files, run shell commands, and drive the desktop (click/type/screenshot) on that machine. It works through the Executor's `computer` connector (the same connectors/discover/describe/call path as every other integration), so there is no separate tunnel client and no token. Load this when the task is about acting ON a specific physical/remote computer the user has connected ("on my laptop…", "read ~/Downloads on my machine", "run this on my desktop", "click the button on my screen"), or when the user asks how the agent reaches their computer. For files INSIDE this sandbox, just use normal shell/fs — not this.

Use the Kortix Executor to reach external systems from a session. Prefer the `kortix executor` CLI for agent work, use `@kortix/executor-sdk` for durable TypeScript workflows and reusable skills, and treat the `kortix executor mcp` server as an optional compatibility face. Load whenever the user asks the agent to act in an external app/API, inspect available connectors/tools, add/configure a connector, or work with `connectors:` in kortix.yaml.